Shopping basket monitoring using computer vision and machine learning

Abstract

A system for monitoring shopping baskets (e.g., baskets on human-propelled carts, motorized carts, or hand-carried baskets) can include a computer vision unit that can image a surveillance region (e.g., an exit to a store), determine whether a basket is empty or loaded with merchandise, and assess a potential for theft of the merchandise. The computer vision unit can include a camera and an image processor programmed to execute a computer vision algorithm to identify shopping baskets and determine a load status of the basket. The computer vision algorithm can comprise a neural network. The system can identify an at least partially loaded shopping basket that is exiting the store, without indicia of having paid for the merchandise, and execute an anti-theft action, e.g., actuating an alarm, notifying store personnel, activating a store surveillance system, activating an anti-theft device associated with the basket (e.g., a locking shopping cart wheel), etc.

1 . (canceled) 
     
     
         2 . An anti-theft system configured to reduce false positives of theft events in a retail store environment, the anti-theft system comprising:
 a wheel configured to connect to a human-propelled shopping cart having a basket for merchandise, the wheel comprising:
 a radio frequency (RF) cart transceiver; and 
 a brake configured to inhibit movement of the cart when the brake is engaged; and 
   a computer vision unit (CVU) comprising:
 a camera configured to be mounted in the retail store and to obtain images of an exit region of the retail store; 
 an image processor; and 
 an RF communication node configured to wirelessly communicate with the RF cart transceiver; 
 the CVU programmed to:
 analyze, with the image processor, the images; 
 identify a non-theft event involving the cart, wherein the identification comprises: (i) a determination that the basket of the cart is empty and (ii) a determination that the cart is attempting to exit the retail store; and 
 in response to the identification of the non-theft event involving the cart, wirelessly communicate using the RF communication node a command to the RF cart transceiver that the cart is authorized to exit the retail store. 
 
   
     
     
         3 . The anti-theft system of  claim 2 , wherein the command to the RF cart transceiver that the cart is authorized to exit the retail store comprises an instruction not to engage the brake. 
     
     
         4 . The anti-theft system of  claim 2 , wherein the determination that the basket of the cart is empty comprises a determination that a child seat of the cart is empty. 
     
     
         5 . The anti-theft system of  claim 2 , wherein the CVU is further programmed to store the images in a remote, non-transitory computer storage medium. 
     
     
         6 . The anti-theft system of  claim 2 , wherein the camera is to be mounted to a ceiling in the retail store. 
     
     
         7 . The anti-theft system of  claim 2 , wherein the camera and the RF communication node are disposed in a housing to be mounted to a wall or ceiling in the retail store and the image processor is disposed remotely from the housing. 
     
     
         8 . The anti-theft system of  claim 2 , wherein the determination that the basket of the cart is empty is performed by a trained machine learning algorithm. 
     
     
         9 . An anti-theft system configured to reduce false positives of theft events in a retail store environment, the anti-theft system comprising:
 a wheel configured to connect to a human-propelled shopping cart having a basket for merchandise, the wheel comprising:
 a radio frequency (RF) cart transceiver; and 
 a brake configured to inhibit movement of the cart when the brake is engaged; 
   a door manager to be mounted in the retail store, the door manager comprising an RF communication node configured to wirelessly communicate with the RF cart transceiver; and   a computer vision unit (CVU) in communication with the door manager, the CVU comprising:
 a camera configured to be mounted in the retail store and to obtain images of an exit region of the retail store; and 
 an image processor; 
 the CVU programmed to:
 analyze, with the image processor, the images; 
 identify a non-theft event involving the cart, wherein the identification comprises: (i) a determination that the basket of the cart is empty and (ii) a determination that the cart is attempting to exit the retail store; and 
 in response to the identification of the non-theft event involving the cart, instruct the door manager to not take an anti-theft action or to wirelessly communicate using the RF communication node an exit authorization command to the RF cart transceiver, thereby allowing the cart to exit the retail store. 
 
   
     
     
         10 . The anti-theft system of  claim 9 , wherein the anti-theft action comprises wirelessly communicating a command to the wheel to engage the brake. 
     
     
         11 . The anti-theft system of  claim 9 , wherein the exit authorization command comprises a command not to engage the brake. 
     
     
         12 . The anti-theft system of  claim 9 , wherein the CVU is further programmed with a trained machine learning algorithm configured to determine a load status of the cart. 
     
     
         13 . The anti-theft system of  claim 9 , wherein the door manager is mounted next to an exit door of the retail store. 
     
     
         14 . An anti-theft system configured to reduce false positives of theft events in a retail store environment, the anti-theft system comprising:
 a wheel configured to connect to a human-propelled shopping cart having a basket for merchandise, the wheel comprising:
 a radio frequency (RF) cart transceiver; and 
 a brake configured to inhibit movement of the cart when the brake is engaged; and 
   a computer vision unit (CVU) comprising:
 a camera configured to be mounted in the retail store and to obtain images of an exit region of the retail store; 
 an image processor; and 
 an RF communication node configured to wirelessly communicate with the RF cart transceiver; 
 the CVU programmed to:
 analyze the images; 
 identify a non-theft event involving the cart, wherein the identification comprises: (i) a determination that the basket of the cart does not contain merchandise and (ii) a determination that the cart is attempting to exit the retail store; and 
 in response to the identification of the non-theft event involving the cart, wirelessly communicate using the RF communication node a command to the RF cart transceiver that the cart is authorized to exit the retail store. 
 
   
     
     
         15 . The anti-theft system of  claim 14 , wherein the CVU is further programmed to determine that the basket contains non-merchandise, the non-merchandise comprising at least one of a child and a handbag. 
     
     
         16 . The anti-theft system of  claim 14  wherein the command to the RF cart transceiver at the cart is authorized to exit the retail store comprises an instruction not to engage the brake. 
     
     
         17 . The anti-theft system of  claim 14 , wherein the determination that the basket of the cart does not contain merchandise comprises a determination that a child seat of the cart does not contain merchandise. 
     
     
         18 . The anti-theft system of  claim 14 , wherein the CVU is further programmed to determine that a child seat of the cart contains non-merchandise, the non-merchandise comprising at least one of a child or a handbag. 
     
     
         19 . The anti-theft system of  claim 14 , wherein the CVU is further programmed to store the images in a remote, non-transitory computer storage medium. 
     
     
         20 . The anti-theft system of  claim 14 , wherein the camera is to be mounted to a ceiling in the retail store. 
     
     
         21 . The anti-theft system of  claim 14 , wherein the camera and the RF communication node are disposed in a housing to be mounted to a wall or ceiling in the retail store and the image processor is disposed remotely from the housing. 
     
     
         22 . The anti-theft system of  claim 14 , wherein the exit region comprises an exit door. 
     
     
         23 . An anti-theft system configured to reduce false positives of theft events in a retail store environment, the anti-theft system comprising:
 a wheel configured to connect to a human-propelled shopping cart having a basket for merchandise, the wheel comprising:
 a radio frequency (RF) cart transceiver; and 
 a brake configured to inhibit movement of the cart when the brake is engaged; 
   a door manager to be mounted in the retail store, the door manager comprising an RF communication node configured to wirelessly communicate with the RF cart transceiver; and   a computer vision unit (CVU) in communication with the door manager, the CVU comprising:
 a camera configured to be mounted in the retail store and to obtain images of an exit region of the retail store; and 
 an image processor; 
 the CVU programmed to:
 analyze the images; 
 identify a non-theft event involving the cart, wherein the identification comprises: (i) a determination that the basket of the cart does not contain merchandise and (ii) a determination that the cart is attempting to exit the retail store; and 
 in response to the identification of the non-theft event involving the cart, instruct the door manager to not take an anti-theft action or to wirelessly communicate using the RF communication node an exit authorization command to the RF cart transceiver, thereby allowing the cart to exit the retail store. 
 
   
     
     
         24 . The anti-theft system of  claim 23 , wherein the CVU is further programmed to determine that the basket contains non-merchandise, the non-merchandise comprising at least one of a child and a handbag. 
     
     
         25 . The anti-theft system of  claim 23 , wherein the anti-theft action comprises wirelessly communicating a command to the wheel to engage the brake. 
     
     
         26 . The anti-theft system of  claim 23 , wherein the exit authorization command comprises a command not to engage the brake. 
     
     
         27 . The anti-theft system of  claim 23 , wherein the determination that the basket of the cart does not contain merchandise comprises a determination that a child seat of the cart does not contain merchandise.